This dramatically reduces the lock contention on systems with slower
(non-TSC) timecounters. With TSC the difference is minimal, but since
this lock is pretty congested, any improvement counts. Plus I don't
see any reason to do it under the lock other than the latency of the
lock itself, which this change actually reduces.
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Closes#12281
This mostly reverts "3537 want pool io kstats" commit of 8 years ago.
From one side this code using pool-wide locks became pretty bad for
performance, creating significant lock contention in I/O pipeline.
From another, there are more efficient ways now to obtain detailed
statistics, while this statistics is illumos-specific and much less
usable on Linux and FreeBSD, reported only via procfs/sysctls.
This commit does not remove KSTAT_TYPE_IO implementation, that may
be removed later together with already unused KSTAT_TYPE_INTR and
KSTAT_TYPE_TIMER.
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Sponsored-By: iXsystems, Inc.
Closes#12212
The `abd_get_offset_*()` routines create an abd_t that references
another abd_t, and doesn't allocate any pages/buffers of its own. In
some workloads, these routines may be called frequently, to create many
abd_t's representing small pieces of a single large abd_t. In
particular, the upcoming RAIDZ Expansion project makes heavy use of
these routines.
This commit adds the ability for the caller to allocate and provide the
abd_t struct to a variant of `abd_get_offset_*()`. This eliminates the
cost of allocating the abd_t and performing the accounting associated
with it (`abdstat_struct_size`). The RAIDZ/DRAID code uses this for
the `rc_abd`, which references the zio's abd. The upcoming RAIDZ
Expansion project will leverage this infrastructure to increase
performance of reads post-expansion by around 50%.
Additionally, some of the interfaces around creating and destroying
abd_t's are cleaned up. Most significantly, the distinction between
`abd_put()` and `abd_free()` is eliminated; all types of abd_t's are
now disposed of with `abd_free()`.
Reviewed-by: Brian Atkinson <batkinson@lanl.gov>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Matthew Ahrens <mahrens@delphix.com>
Issue #8853Closes#11439
Since 8c4fb36a24 (PR #7795) spa_has_pending_synctask() started to
take two more locks per write inside txg_all_lists_empty(). I am
surprised those pool-wide locks are not contended, but still their
operations are visible in CPU profiles under contended vdev lock.
This commit slightly changes vdev_queue_max_async_writes() flow to
not call the function if we are going to return max_active any way
due to high amount of dirty data. It allows to save some CPU time
exactly when the pool is busy.
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-By: Tom Caputi <caputit1@tcnj.edu>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Closes#11280
It was found that setting min_active tunables for non-interactive I/Os
makes them stuck. It is caused by zfs_vdev_nia_delay, that can never
be reached if we never issue any I/Os due to min_active set to zero.
Fix this by issuing at least one non-interactive I/O at a time when
there are no interactive I/Os. When there are interactive I/Os, zero
min_active allows to completely block any non-interactive I/O. It may
min_active starvation in some scenarios, but who we are to deny foot
shooting?
Reviewed-by: George Melikov <mail@gmelikov.ru>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Closes#11261
Investigating influence of scrub (especially sequential) on random read
latency I've noticed that on some HDDs single 4KB read may take up to 4
seconds! Deeper investigation shown that many HDDs heavily prioritize
sequential reads even when those are submitted with queue depth of 1.
This patch addresses the latency from two sides:
- by using _min_active queue depths for non-interactive requests while
the interactive request(s) are active and few requests after;
- by throttling it further if no interactive requests has completed
while configured amount of non-interactive did.
While there, I've also modified vdev_queue_class_to_issue() to give
more chances to schedule at least _min_active requests to the lowest
priorities. It should reduce starvation if several non-interactive
processes are running same time with some interactive and I think should
make possible setting of zfs_vdev_max_active to as low as 1.
I've benchmarked this change with 4KB random reads from ZVOL with 16KB
block size on newly written non-fragmented pool. On fragmented pool I
also saw improvements, but not so dramatic. Below are log2 histograms
of the random read latency in milliseconds for different devices:
4 2x mirror vdevs of SATA HDD WDC WD20EFRX-68EUZN0 before:
0, 0, 2, 1, 12, 21, 19, 18, 10, 15, 17, 21
after:
0, 0, 0, 24, 101, 195, 419, 250, 47, 4, 0, 0
, that means maximum latency reduction from 2s to 500ms.
4 2x mirror vdevs of SATA HDD WDC WD80EFZX-68UW8N0 before:
0, 0, 2, 31, 38, 28, 18, 12, 17, 20, 24, 10, 3
after:
0, 0, 55, 247, 455, 470, 412, 181, 36, 0, 0, 0, 0
, i.e. from 4s to 250ms.
1 SAS HDD SEAGATE ST14000NM0048 before:
0, 0, 29, 70, 107, 45, 27, 1, 0, 0, 1, 4, 19
after:
1, 29, 681, 1261, 676, 1633, 67, 1, 0, 0, 0, 0, 0
, i.e. from 4s to 125ms.
1 SAS SSD SEAGATE XS3840TE70014 before (microseconds):
0, 0, 0, 0, 0, 0, 0, 0, 70, 18343, 82548, 618
after:
0, 0, 0, 0, 0, 0, 0, 0, 283, 92351, 34844, 90
I've also measured scrub time during the test and on idle pools. On
idle fragmented pool I've measured scrub getting few percent faster
due to use of QD3 instead of QD2 before. On idle non-fragmented pool
I've measured no difference. On busy non-fragmented pool I've measured
scrub time increase about 1.5-1.7x, while IOPS increase reached 5-9x.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Sponsored-By: iXsystems, Inc.
Closes#11166
This patch adds a new top-level vdev type called dRAID, which stands
for Distributed parity RAID. This pool configuration allows all dRAID
vdevs to participate when rebuilding to a distributed hot spare device.
This can substantially reduce the total time required to restore full
parity to pool with a failed device.
A dRAID pool can be created using the new top-level `draid` type.
Like `raidz`, the desired redundancy is specified after the type:
`draid[1,2,3]`. No additional information is required to create the
pool and reasonable default values will be chosen based on the number
of child vdevs in the dRAID vdev.
zpool create <pool> draid[1,2,3] <vdevs...>
Unlike raidz, additional optional dRAID configuration values can be
provided as part of the draid type as colon separated values. This
allows administrators to fully specify a layout for either performance
or capacity reasons. The supported options include:
zpool create <pool> \
draid[<parity>][:<data>d][:<children>c][:<spares>s] \
<vdevs...>
- draid[parity] - Parity level (default 1)
- draid[:<data>d] - Data devices per group (default 8)
- draid[:<children>c] - Expected number of child vdevs
- draid[:<spares>s] - Distributed hot spares (default 0)
Abbreviated example `zpool status` output for a 68 disk dRAID pool
with two distributed spares using special allocation classes.
```
pool: tank
state: ONLINE
config:
NAME STATE READ WRITE CKSUM
slag7 ONLINE 0 0 0
draid2:8d:68c:2s-0 ONLINE 0 0 0
L0 ONLINE 0 0 0
L1 ONLINE 0 0 0
...
U25 ONLINE 0 0 0
U26 ONLINE 0 0 0
spare-53 ONLINE 0 0 0
U27 ONLINE 0 0 0
draid2-0-0 ONLINE 0 0 0
U28 ONLINE 0 0 0
U29 ONLINE 0 0 0
...
U42 ONLINE 0 0 0
U43 ONLINE 0 0 0
special
mirror-1 ONLINE 0 0 0
L5 ONLINE 0 0 0
U5 ONLINE 0 0 0
mirror-2 ONLINE 0 0 0
L6 ONLINE 0 0 0
U6 ONLINE 0 0 0
spares
draid2-0-0 INUSE currently in use
draid2-0-1 AVAIL
```
When adding test coverage for the new dRAID vdev type the following
options were added to the ztest command. These options are leverages
by zloop.sh to test a wide range of dRAID configurations.
-K draid|raidz|random - kind of RAID to test
-D <value> - dRAID data drives per group
-S <value> - dRAID distributed hot spares
-R <value> - RAID parity (raidz or dRAID)
The zpool_create, zpool_import, redundancy, replacement and fault
test groups have all been updated provide test coverage for the
dRAID feature.
Co-authored-by: Isaac Huang <he.huang@intel.com>
Co-authored-by: Mark Maybee <mmaybee@cray.com>
Co-authored-by: Don Brady <don.brady@delphix.com>
Co-authored-by: Matthew Ahrens <mahrens@delphix.com>
Co-authored-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Mark Maybee <mmaybee@cray.com>
Reviewed-by: Matt Ahrens <matt@delphix.com>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#10102
The device_rebuild feature enables sequential reconstruction when
resilvering. Mirror vdevs can be rebuilt in LBA order which may
more quickly restore redundancy depending on the pools average block
size, overall fragmentation and the performance characteristics
of the devices. However, block checksums cannot be verified
as part of the rebuild thus a scrub is automatically started after
the sequential resilver completes.
The new '-s' option has been added to the `zpool attach` and
`zpool replace` command to request sequential reconstruction
instead of healing reconstruction when resilvering.
zpool attach -s <pool> <existing vdev> <new vdev>
zpool replace -s <pool> <old vdev> <new vdev>
The `zpool status` output has been updated to report the progress
of sequential resilvering in the same way as healing resilvering.
The one notable difference is that multiple sequential resilvers
may be in progress as long as they're operating on different
top-level vdevs.
The `zpool wait -t resilver` command was extended to wait on
sequential resilvers. From this perspective they are no different
than healing resilvers.
Sequential resilvers cannot be supported for RAIDZ, but are
compatible with the dRAID feature being developed.
As part of this change the resilver_restart_* tests were moved
in to the functional/replacement directory. Additionally, the
replacement tests were renamed and extended to verify both
resilvering and rebuilding.
Original-patch-by: Isaac Huang <he.huang@intel.com>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: John Poduska <jpoduska@datto.com>
Co-authored-by: Mark Maybee <mmaybee@cray.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#10349
Mark functions used only in the same translation unit as static. This
only includes functions that do not have a prototype in a header file
either.
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Arvind Sankar <nivedita@alum.mit.edu>
Closes#10470
Adding the gang ABD type, which allows for linear and scatter ABDs to
be chained together into a single ABD.
This can be used to avoid doing memory copies to/from ABDs. An example
of this can be found in vdev_queue.c in the vdev_queue_aggregate()
function.
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Co-authored-by: Brian <bwa@clemson.edu>
Co-authored-by: Mark Maybee <mmaybee@cray.com>
Signed-off-by: Brian Atkinson <batkinson@lanl.gov>
Closes#10069
This patch implements a new tree structure for ZFS, and uses it to
store range trees more efficiently.
The new structure is approximately a B-tree, though there are some
small differences from the usual characterizations. The tree has core
nodes and leaf nodes; each contain data elements, which the elements
in the core nodes acting as separators between its children. The
difference between core and leaf nodes is that the core nodes have an
array of children, while leaf nodes don't. Every node in the tree may
be only partially full; in most cases, they are all at least 50% full
(in terms of element count) except for the root node, which can be
less full. Underfull nodes will steal from their neighbors or merge to
remain full enough, while overfull nodes will split in two. The data
elements are contained in tree-controlled buffers; they are copied
into these on insertion, and overwritten on deletion. This means that
the elements are not independently allocated, which reduces overhead,
but also means they can't be shared between trees (and also that
pointers to them are only valid until a side-effectful tree operation
occurs). The overhead varies based on how dense the tree is, but is
usually on the order of about 50% of the element size; the per-node
overheads are very small, and so don't make a significant difference.
The trees can accept arbitrary records; they accept a size and a
comparator to allow them to be used for a variety of purposes.
The new trees replace the AVL trees used in the range trees today.
Currently, the range_seg_t structure contains three 8 byte integers
of payload and two 24 byte avl_tree_node_ts to handle its storage in
both an offset-sorted tree and a size-sorted tree (total size: 64
bytes). In the new model, the range seg structures are usually two 4
byte integers, but a separate one needs to exist for the size-sorted
and offset-sorted tree. Between the raw size, the 50% overhead, and
the double storage, the new btrees are expected to use 8*1.5*2 = 24
bytes per record, or 33.3% as much memory as the AVL trees (this is
for the purposes of storing metaslab range trees; for other purposes,
like scrubs, they use ~50% as much memory).
We reduced the size of the payload in the range segments by teaching
range trees about starting offsets and shifts; since metaslabs have a
fixed starting offset, and they all operate in terms of disk sectors,
we can store the ranges using 4-byte integers as long as the size of
the metaslab divided by the sector size is less than 2^32. For 512-byte
sectors, this is a 2^41 (or 2TB) metaslab, which with the default
settings corresponds to a 256PB disk. 4k sector disks can handle
metaslabs up to 2^46 bytes, or 2^63 byte disks. Since we do not
anticipate disks of this size in the near future, there should be
almost no cases where metaslabs need 64-byte integers to store their
ranges. We do still have the capability to store 64-byte integer ranges
to account for cases where we are storing per-vdev (or per-dnode) trees,
which could reasonably go above the limits discussed. We also do not
store fill information in the compact version of the node, since it
is only used for sorted scrub.
We also optimized the metaslab loading process in various other ways
to offset some inefficiencies in the btree model. While individual
operations (find, insert, remove_from) are faster for the btree than
they are for the avl tree, remove usually requires a find operation,
while in the AVL tree model the element itself suffices. Some clever
changes actually caused an overall speedup in metaslab loading; we use
approximately 40% less cpu to load metaslabs in our tests on Illumos.
Another memory and performance optimization was achieved by changing
what is stored in the size-sorted trees. When a disk is heavily
fragmented, the df algorithm used by default in ZFS will almost always
find a number of small regions in its initial cursor-based search; it
will usually only fall back to the size-sorted tree to find larger
regions. If we increase the size of the cursor-based search slightly,
and don't store segments that are smaller than a tunable size floor
in the size-sorted tree, we can further cut memory usage down to
below 20% of what the AVL trees store. This also results in further
reductions in CPU time spent loading metaslabs.
The 16KiB size floor was chosen because it results in substantial memory
usage reduction while not usually resulting in situations where we can't
find an appropriate chunk with the cursor and are forced to use an
oversized chunk from the size-sorted tree. In addition, even if we do
have to use an oversized chunk from the size-sorted tree, the chunk
would be too small to use for ZIL allocations, so it isn't as big of a
loss as it might otherwise be. And often, more small allocations will
follow the initial one, and the cursor search will now find the
remainder of the chunk we didn't use all of and use it for subsequent
allocations. Practical testing has shown little or no change in
fragmentation as a result of this change.
If the size-sorted tree becomes empty while the offset sorted one still
has entries, it will load all the entries from the offset sorted tree
and disregard the size floor until it is unloaded again. This operation
occurs rarely with the default setting, only on incredibly thoroughly
fragmented pools.
There are some other small changes to zdb to teach it to handle btrees,
but nothing major.
Reviewed-by: George Wilson <gwilson@delphix.com>
Reviewed-by: Matt Ahrens <matt@delphix.com>
Reviewed by: Sebastien Roy seb@delphix.com
Reviewed-by: Igor Kozhukhov <igor@dilos.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Paul Dagnelie <pcd@delphix.com>
Closes#9181
Adds ZFS_MODULE_PARAM to abstract module parameter
setting to operating systems other than Linux.
Reviewed-by: Jorgen Lundman <lundman@lundman.net>
Reviewed-by: Igor Kozhukhov <igor@dilos.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Matt Macy <mmacy@FreeBSD.org>
Signed-off-by: Ryan Moeller <ryan@ixsystems.com>
Closes#9230
Reviewed-by: Matt Ahrens <matt@delphix.com>
Reviewed-by: Ryan Moeller <ryan@ixsystems.com>
Reviewed-by: Richard Laager <rlaager@wiktel.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Andrea Gelmini <andrea.gelmini@gelma.net>
Closes#9240
Memory copy is too heavy operation to do under the congested lock.
Moving it out reduces congestion by many times to almost invisible.
Since the original zio removed from the queue, and the child zio is
not executed yet, I don't see why would the copy need protection.
My guess it just remained like this from the time when lock was not
dropped here, which was added later to fix lock ordering issue.
Multi-threaded sequential write tests with both HDD and SSD pools
with ZVOL block sizes of 4KB, 16KB, 64KB and 128KB all show major
reduction of lock congestion, saving from 15% to 35% of CPU time
and increasing throughput from 10% to 40%.
Reviewed-by: Richard Yao <ryao@gentoo.org>
Reviewed-by: Matt Ahrens <mahrens@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Closes#8890
UNMAP/TRIM support is a frequently-requested feature to help
prevent performance from degrading on SSDs and on various other
SAN-like storage back-ends. By issuing UNMAP/TRIM commands for
sectors which are no longer allocated the underlying device can
often more efficiently manage itself.
This TRIM implementation is modeled on the `zpool initialize`
feature which writes a pattern to all unallocated space in the
pool. The new `zpool trim` command uses the same vdev_xlate()
code to calculate what sectors are unallocated, the same per-
vdev TRIM thread model and locking, and the same basic CLI for
a consistent user experience. The core difference is that
instead of writing a pattern it will issue UNMAP/TRIM commands
for those extents.
The zio pipeline was updated to accommodate this by adding a new
ZIO_TYPE_TRIM type and associated spa taskq. This new type makes
is straight forward to add the platform specific TRIM/UNMAP calls
to vdev_disk.c and vdev_file.c. These new ZIO_TYPE_TRIM zios are
handled largely the same way as ZIO_TYPE_READs or ZIO_TYPE_WRITEs.
This makes it possible to largely avoid changing the pipieline,
one exception is that TRIM zio's may exceed the 16M block size
limit since they contain no data.
In addition to the manual `zpool trim` command, a background
automatic TRIM was added and is controlled by the 'autotrim'
property. It relies on the exact same infrastructure as the
manual TRIM. However, instead of relying on the extents in a
metaslab's ms_allocatable range tree, a ms_trim tree is kept
per metaslab. When 'autotrim=on', ranges added back to the
ms_allocatable tree are also added to the ms_free tree. The
ms_free tree is then periodically consumed by an autotrim
thread which systematically walks a top level vdev's metaslabs.
Since the automatic TRIM will skip ranges it considers too small
there is value in occasionally running a full `zpool trim`. This
may occur when the freed blocks are small and not enough time
was allowed to aggregate them. An automatic TRIM and a manual
`zpool trim` may be run concurrently, in which case the automatic
TRIM will yield to the manual TRIM.
Reviewed-by: Jorgen Lundman <lundman@lundman.net>
Reviewed-by: Tim Chase <tim@chase2k.com>
Reviewed-by: Matt Ahrens <mahrens@delphix.com>
Reviewed-by: George Wilson <george.wilson@delphix.com>
Reviewed-by: Serapheim Dimitropoulos <serapheim@delphix.com>
Contributions-by: Saso Kiselkov <saso.kiselkov@nexenta.com>
Contributions-by: Tim Chase <tim@chase2k.com>
Contributions-by: Chunwei Chen <tuxoko@gmail.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#8419Closes#598
Before sequential scrub patches ZFS never aggregated I/Os above 128KB.
Sequential scrub bumped that to 1MB, supposedly to reduce number of
head seeks for spinning disks. But for SSDs it makes little to no
sense, especially on FreeBSD, where due to MAXPHYS limitation device
will likely still see bunch of 128KB I/Os instead of one large.
Having more strict aggregation limit for SSDs allows to avoid
allocation of large memory buffer and copy to/from it, that is a
serious problem when throughput reaches gigabytes per second.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Richard Elling <Richard.Elling@RichardElling.com>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Closes#8494
PROBLEM
========
The first access to a block incurs a performance penalty on some platforms
(e.g. AWS's EBS, VMware VMDKs). Therefore we recommend that volumes are
"thick provisioned", where supported by the platform (VMware). This can
create a large delay in getting a new virtual machines up and running (or
adding storage to an existing Engine). If the thick provision step is
omitted, write performance will be suboptimal until all blocks on the LUN
have been written.
SOLUTION
=========
This feature introduces a way to 'initialize' the disks at install or in the
background to make sure we don't incur this first read penalty.
When an entire LUN is added to ZFS, we make all space available immediately,
and allow ZFS to find unallocated space and zero it out. This works with
concurrent writes to arbitrary offsets, ensuring that we don't zero out
something that has been (or is in the middle of being) written. This scheme
can also be applied to existing pools (affecting only free regions on the
vdev). Detailed design:
- new subcommand:zpool initialize [-cs] <pool> [<vdev> ...]
- start, suspend, or cancel initialization
- Creates new open-context thread for each vdev
- Thread iterates through all metaslabs in this vdev
- Each metaslab:
- select a metaslab
- load the metaslab
- mark the metaslab as being zeroed
- walk all free ranges within that metaslab and translate
them to ranges on the leaf vdev
- issue a "zeroing" I/O on the leaf vdev that corresponds to
a free range on the metaslab we're working on
- continue until all free ranges for this metaslab have been
"zeroed"
- reset/unmark the metaslab being zeroed
- if more metaslabs exist, then repeat above tasks.
- if no more metaslabs, then we're done.
- progress for the initialization is stored on-disk in the vdev’s
leaf zap object. The following information is stored:
- the last offset that has been initialized
- the state of the initialization process (i.e. active,
suspended, or canceled)
- the start time for the initialization
- progress is reported via the zpool status command and shows
information for each of the vdevs that are initializing
Porting notes:
- Added zfs_initialize_value module parameter to set the pattern
written by "zpool initialize".
- Added zfs_vdev_{initializing,removal}_{min,max}_active module options.
Authored by: George Wilson <george.wilson@delphix.com>
Reviewed by: John Wren Kennedy <john.kennedy@delphix.com>
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: Pavel Zakharov <pavel.zakharov@delphix.com>
Reviewed by: Prakash Surya <prakash.surya@delphix.com>
Reviewed by: loli10K <ezomori.nozomu@gmail.com>
Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov>
Approved by: Richard Lowe <richlowe@richlowe.net>
Signed-off-by: Tim Chase <tim@chase2k.com>
Ported-by: Tim Chase <tim@chase2k.com>
OpenZFS-issue: https://www.illumos.org/issues/9102
OpenZFS-commit: https://github.com/openzfs/openzfs/commit/c3963210ebCloses#8230
There are some issues with the way the seq_file interface is implemented
for kstats backed by linked lists (zfs_dbgmsgs and certain per-pool
debugging info):
* We don't account for the fact that seq_file sometimes visits a node
multiple times, which results in missing messages when read through
procfs.
* We don't keep separate state for each reader of a file, so concurrent
readers will receive incorrect results.
* We don't account for the fact that entries may have been removed from
the list between read syscalls, so reading from these files in procfs
can cause the system to crash.
This change fixes these issues and adds procfs_list, a wrapper around a
linked list which abstracts away the details of implementing the
seq_file interface for a list and exposing the contents of the list
through procfs.
Reviewed by: Don Brady <don.brady@delphix.com>
Reviewed-by: Serapheim Dimitropoulos <serapheim@delphix.com>
Reviewed by: Brad Lewis <brad.lewis@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: John Gallagher <john.gallagher@delphix.com>
External-issue: LX-1211
Closes#7819
Overview
========
We parallelize the allocation process by creating the concept of
"allocators". There are a certain number of allocators per metaslab
group, defined by the value of a tunable at pool open time. Each
allocator for a given metaslab group has up to 2 active metaslabs; one
"primary", and one "secondary". The primary and secondary weight mean
the same thing they did in in the pre-allocator world; primary metaslabs
are used for most allocations, secondary metaslabs are used for ditto
blocks being allocated in the same metaslab group. There is also the
CLAIM weight, which has been separated out from the other weights, but
that is less important to understanding the patch. The active metaslabs
for each allocator are moved from their normal place in the metaslab
tree for the group to the back of the tree. This way, they will not be
selected for use by other allocators searching for new metaslabs unless
all the passive metaslabs are unsuitable for allocations. If that does
happen, the allocators will "steal" from each other to ensure that IOs
don't fail until there is truly no space left to perform allocations.
In addition, the alloc queue for each metaslab group has been broken
into a separate queue for each allocator. We don't want to dramatically
increase the number of inflight IOs on low-end systems, because it can
significantly increase txg times. On the other hand, we want to ensure
that there are enough IOs for each allocator to allow for good
coalescing before sending the IOs to the disk. As a result, we take a
compromise path; each allocator's alloc queue max depth starts at a
certain value for every txg. Every time an IO completes, we increase the
max depth. This should hopefully provide a good balance between the two
failure modes, while not dramatically increasing complexity.
We also parallelize the spa_alloc_tree and spa_alloc_lock, which cause
very similar contention when selecting IOs to allocate. This
parallelization uses the same allocator scheme as metaslab selection.
Performance Results
===================
Performance improvements from this change can vary significantly based
on the number of CPUs in the system, whether or not the system has a
NUMA architecture, the speed of the drives, the values for the various
tunables, and the workload being performed. For an fio async sequential
write workload on a 24 core NUMA system with 256 GB of RAM and 8 128 GB
SSDs, there is a roughly 25% performance improvement.
Future Work
===========
Analysis of the performance of the system with this patch applied shows
that a significant new bottleneck is the vdev disk queues, which also
need to be parallelized. Prototyping of this change has occurred, and
there was a performance improvement, but more work needs to be done
before its stability has been verified and it is ready to be upstreamed.
Authored by: Paul Dagnelie <pcd@delphix.com>
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Serapheim Dimitropoulos <serapheim.dimitro@delphix.com>
Reviewed by: Alexander Motin <mav@FreeBSD.org>
Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov>
Approved by: Gordon Ross <gwr@nexenta.com>
Ported-by: Paul Dagnelie <pcd@delphix.com>
Signed-off-by: Paul Dagnelie <pcd@delphix.com>
Porting Notes:
* Fix reservation test failures by increasing tolerance.
OpenZFS-issue: https://illumos.org/issues/9112
OpenZFS-commit: https://github.com/openzfs/openzfs/commit/3f3cc3c3Closes#7682
This fixes an assert in vdev_queue_change_io_priority():
VERIFY3(zio->io_priority < ZIO_PRIORITY_NUM_QUEUEABLE) failed (7 < 6)
PANIC at vdev_queue.c:832:vdev_queue_change_io_priority()
Reviewed-by: Tom Caputi <tcaputi@datto.com>
Reviewed-by: George Melikov <mail@gmelikov.ru>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Tony Hutter <hutter2@llnl.gov>
Closes#7566Closes#7542
Minimal changes required to integrate the SPL sources in to the
ZFS repository build infrastructure and packaging.
Build system and packaging:
* Renamed SPL_* autoconf m4 macros to ZFS_*.
* Removed redundant SPL_* autoconf m4 macros.
* Updated the RPM spec files to remove SPL package dependency.
* The zfs package obsoletes the spl package, and the zfs-kmod
package obsoletes the spl-kmod package.
* The zfs-kmod-devel* packages were updated to add compatibility
symlinks under /usr/src/spl-x.y.z until all dependent packages
can be updated. They will be removed in a future release.
* Updated copy-builtin script for in-kernel builds.
* Updated DKMS package to include the spl.ko.
* Updated stale AUTHORS file to include all contributors.
* Updated stale COPYRIGHT and included the SPL as an exception.
* Renamed README.markdown to README.md
* Renamed OPENSOLARIS.LICENSE to LICENSE.
* Renamed DISCLAIMER to NOTICE.
Required code changes:
* Removed redundant HAVE_SPL macro.
* Removed _BOOT from nvpairs since it doesn't apply for Linux.
* Initial header cleanup (removal of empty headers, refactoring).
* Remove SPL repository clone/build from zimport.sh.
* Use of DEFINE_RATELIMIT_STATE and DEFINE_SPINLOCK removed due
to build issues when forcing C99 compilation.
* Replaced legacy ACCESS_ONCE with READ_ONCE.
* Include needed headers for `current` and `EXPORT_SYMBOL`.
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: Olaf Faaland <faaland1@llnl.gov>
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Pavel Zakharov <pavel.zakharov@delphix.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
TEST_ZIMPORT_SKIP="yes"
Closes#7556
OpenZFS 7614 - zfs device evacuation/removal
OpenZFS 9064 - remove_mirror should wait for device removal to complete
This project allows top-level vdevs to be removed from the storage pool
with "zpool remove", reducing the total amount of storage in the pool.
This operation copies all allocated regions of the device to be removed
onto other devices, recording the mapping from old to new location.
After the removal is complete, read and free operations to the removed
(now "indirect") vdev must be remapped and performed at the new location
on disk. The indirect mapping table is kept in memory whenever the pool
is loaded, so there is minimal performance overhead when doing operations
on the indirect vdev.
The size of the in-memory mapping table will be reduced when its entries
become "obsolete" because they are no longer used by any block pointers
in the pool. An entry becomes obsolete when all the blocks that use
it are freed. An entry can also become obsolete when all the snapshots
that reference it are deleted, and the block pointers that reference it
have been "remapped" in all filesystems/zvols (and clones). Whenever an
indirect block is written, all the block pointers in it will be "remapped"
to their new (concrete) locations if possible. This process can be
accelerated by using the "zfs remap" command to proactively rewrite all
indirect blocks that reference indirect (removed) vdevs.
Note that when a device is removed, we do not verify the checksum of
the data that is copied. This makes the process much faster, but if it
were used on redundant vdevs (i.e. mirror or raidz vdevs), it would be
possible to copy the wrong data, when we have the correct data on e.g.
the other side of the mirror.
At the moment, only mirrors and simple top-level vdevs can be removed
and no removal is allowed if any of the top-level vdevs are raidz.
Porting Notes:
* Avoid zero-sized kmem_alloc() in vdev_compact_children().
The device evacuation code adds a dependency that
vdev_compact_children() be able to properly empty the vdev_child
array by setting it to NULL and zeroing vdev_children. Under Linux,
kmem_alloc() and related functions return a sentinel pointer rather
than NULL for zero-sized allocations.
* Remove comment regarding "mpt" driver where zfs_remove_max_segment
is initialized to SPA_MAXBLOCKSIZE.
Change zfs_condense_indirect_commit_entry_delay_ticks to
zfs_condense_indirect_commit_entry_delay_ms for consistency with
most other tunables in which delays are specified in ms.
* ZTS changes:
Use set_tunable rather than mdb
Use zpool sync as appropriate
Use sync_pool instead of sync
Kill jobs during test_removal_with_operation to allow unmount/export
Don't add non-disk names such as "mirror" or "raidz" to $DISKS
Use $TEST_BASE_DIR instead of /tmp
Increase HZ from 100 to 1000 which is more common on Linux
removal_multiple_indirection.ksh
Reduce iterations in order to not time out on the code
coverage builders.
removal_resume_export:
Functionally, the test case is correct but there exists a race
where the kernel thread hasn't been fully started yet and is
not visible. Wait for up to 1 second for the removal thread
to be started before giving up on it. Also, increase the
amount of data copied in order that the removal not finish
before the export has a chance to fail.
* MMP compatibility, the concept of concrete versus non-concrete devices
has slightly changed the semantics of vdev_writeable(). Update
mmp_random_leaf_impl() accordingly.
* Updated dbuf_remap() to handle the org.zfsonlinux:large_dnode pool
feature which is not supported by OpenZFS.
* Added support for new vdev removal tracepoints.
* Test cases removal_with_zdb and removal_condense_export have been
intentionally disabled. When run manually they pass as intended,
but when running in the automated test environment they produce
unreliable results on the latest Fedora release.
They may work better once the upstream pool import refectoring is
merged into ZoL at which point they will be re-enabled.
Authored by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Alex Reece <alex@delphix.com>
Reviewed-by: George Wilson <george.wilson@delphix.com>
Reviewed-by: John Kennedy <john.kennedy@delphix.com>
Reviewed-by: Prakash Surya <prakash.surya@delphix.com>
Reviewed by: Richard Laager <rlaager@wiktel.com>
Reviewed by: Tim Chase <tim@chase2k.com>
Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov>
Approved by: Garrett D'Amore <garrett@damore.org>
Ported-by: Tim Chase <tim@chase2k.com>
Signed-off-by: Tim Chase <tim@chase2k.com>
OpenZFS-issue: https://www.illumos.org/issues/7614
OpenZFS-commit: https://github.com/openzfs/openzfs/commit/f539f1ebCloses#6900
In vdev_queue_aggregate() the zio_execute() bypass should not be
called under the vdev queue lock. This can result in a deadlock
as shown in the stack traces below.
Drop the vdev queue lock then walk the parents of the aggregate IO
to determine the list of component IOs to be bypassed. This can
be done safely without holding the io_lock since the new aggregate
IO has not yet been returned and its parents cannot change.
--- THREAD 1 ---
arc_read()
zio_nowait()
zio_vdev_io_start()
vdev_queue_io() <--- mutex_enter(vq->vq_lock)
vdev_queue_io_to_issue()
vdev_queue_aggregate()
zio_execute()
zio_vdev_io_assess()
zio_wait_for_children() <- mutex_enter(zio->io_lock)
--- THREAD 2 --- (inverse order)
arc_read()
zio_change_priority() <- mutex_enter(zio->zio_lock)
vdev_queue_change_io_priority() <- mutex_enter(vq->vq_lock)
Reviewed-by: Tom Caputi <tcaputi@datto.com>
Reviewed-by: Don Brady <don.brady@delphix.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#7307
When sequential scrubs were merged, all calls to arc_read()
(including prefetch IOs) were given ZIO_PRIORITY_ASYNC_READ.
Unfortunately, this behaves badly with an existing issue where
prefetch IOs cannot be re-prioritized after the issue. The
result is that synchronous reads end up in the same vdev_queue
as the scrub IOs and can have (in some workloads) multiple
seconds of latency.
This patch incorporates 2 changes. The first ensures that all
scrub IOs are given ZIO_PRIORITY_SCRUB to allow the vdev_queue
code to differentiate between these I/Os and user prefetches.
Second, this patch introduces zio_change_priority() to provide
the missing capability to upgrade a zio's priority.
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Tom Caputi <tcaputi@datto.com>
Closes#6921Closes#6926
Currently, scrubs and resilvers can take an extremely
long time to complete. This is largely due to the fact
that zfs scans process pools in logical order, as
determined by each block's bookmark. This makes sense
from a simplicity perspective, but blocks in zfs are
often scattered randomly across disks, particularly
due to zfs's copy-on-write mechanisms.
This patch improves performance by splitting scrubs
and resilvers into a metadata scanning phase and an IO
issuing phase. The metadata scan reads through the
structure of the pool and gathers an in-memory queue
of I/Os, sorted by size and offset on disk. The issuing
phase will then issue the scrub I/Os as sequentially as
possible, greatly improving performance.
This patch also updates and cleans up some of the scan
code which has not been updated in several years.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Authored-by: Saso Kiselkov <saso.kiselkov@nexenta.com>
Authored-by: Alek Pinchuk <apinchuk@datto.com>
Authored-by: Tom Caputi <tcaputi@datto.com>
Signed-off-by: Tom Caputi <tcaputi@datto.com>
Closes#3625Closes#6256
With PR 5756 the zfs module now supports c99 and the
remaining past c89 workarounds can be undone.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: George Melikov <mail@gmelikov.ru>
Signed-off-by: Don Brady <don.brady@delphix.com>
Closes#6816
vdev_queue:
- Track the last position of each vdev, including the io size,
in order to detect linear access of the following zio.
- Remove duplicate `vq_lastoffset`
vdev_mirror:
- Correctly calculate the zio offset (signedness issue)
- Deprecate `vdev_queue_register_lastoffset()`
- Add `VDEV_LABEL_START_SIZE` to zio offset of leaf vdevs
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Gvozden Neskovic <neskovic@gmail.com>
Closes#6461
Commit 8542ef8 allowed optional IOs to be aggregated beyond
the specified aggregation limit. Since the aggregation limit
was also used to enforce the maximum block size, setting
`zfs_vdev_aggregation_limit=16777216` could result in an
attempt to allocate an ABD larger than 16M.
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: George Melikov <mail@gmelikov.ru>
Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#6259Closes#6270
Resilver operations frequently cause only a small amount of dirty data
to be written to disk at a time, resulting in the IO scheduler to only
issue 1 write at a time to the resilvering disk. When it is rotational
media the drive will often travel past the next sector to be written
before receiving a write command from ZFS, significantly delaying the
write of the next sector.
Raise zfs_vdev_async_write_min_active so that drives are kept fed
during resilvering.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: DHE <git@dehacked.net>
Issue #4825Closes#5926
Authored by: Matt Ahrens <mahrens@delphix.com>
Reviewed by: Saso Kiselkov <saso.kiselkov@nexenta.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: George Melikov <mail@gmelikov.ru>
Reviewed-by: Don Brady <don.brady@intel.com>
Ported-by: Matt Ahrens <mahrens@delphix.com>
RAID-Z requires that space be allocated in multiples of P+1 sectors,
because this is the minimum size block that can have the required amount
of parity. Thus blocks on RAIDZ1 must be allocated in a multiple of 2
sectors; on RAIDZ2 multiple of 3; and on RAIDZ3 multiple of 4. A sector
is a unit of 2^ashift bytes, typically 512B or 4KB.
To satisfy this constraint, the allocation size is rounded up to the
proper multiple, resulting in up to 3 "pad sectors" at the end of some
blocks. The contents of these pad sectors are not used, so we do not
need to read or write these sectors. However, some storage hardware
performs much worse (around 1/2 as fast) on mostly-contiguous writes
when there are small gaps of non-overwritten data between the writes.
Therefore, ZFS creates "optional" zio's when writing RAID-Z blocks that
include pad sectors. If writing a pad sector will fill the gap between
two (required) writes, we will issue the optional zio, thus doubling
performance. The gap-filling performance improvement was introduced in
July 2009.
Writing the optional zio is done by the io aggregation code in
vdev_queue.c. The problem is that it is also subject to the limit on
the size of aggregate writes, zfs_vdev_aggregation_limit, which is by
default 128KB. For a given block, if the amount of data plus padding
written to a leaf device exceeds zfs_vdev_aggregation_limit, the
optional zio will not be written, resulting in a ~2x performance
degradation.
The problem occurs only for certain values of ashift, compressed block
size, and RAID-Z configuration (number of parity and data disks). It
cannot occur with the default recordsize=128KB. If compression is
enabled, all configurations with recordsize=1MB or larger will be
impacted to some degree.
The problem notably occurs with recordsize=1MB, compression=off, with 10
disks in a RAIDZ2 or RAIDZ3 group (with 512B or 4KB sectors). Therefore
this problem has been known as "the 1MB 10-wide RAIDZ2 (or 3) problem".
The problem also occurs with the following configurations:
With recordsize=512KB or 256KB, compression=off, the problem occurs only
in rarely-used configurations:
* 4-wide RAIDZ1 with recordsize=512KB and ashift=12 (4KB sectors)
* 4-wide RAIDZ2 (either recordsize, either ashift)
* 5-wide RAIDZ2 with recordsize=512KB (either ashift)
* 6-wide RAIDZ2 with recordsize=512KB (either ashift)
With recordsize=1MB, compression=off, ashift=9 (512B sectors)
* RAIDZ1 with 4 or 8 disks
* RAIDZ2 with 4, 8, or 10 disks
* RAIDZ3 with 6, 8, 9, or 10 disks
With recordsize=1MB, compression=off, ashift=12 (4KB sectors)
* RAIDZ1 with 7 or 8 disks
* RAIDZ2 with 4, 5, or 10 disks
* RAIDZ3 with 6, 9, or 10 disks
With recordsize=2MB and larger (which can only be selected by changing
kernel tunables), many configurations are affected, including with
higher numbers of disks (up to 18 disks with recordsize=2MB).
Increase zfs_vdev_aggregation_limit to allow the optional zio to be
aggregated, thus eliminating the problem. Setting it to 256KB fixes all
commonly-used configurations.
The solution is to aggregate optional zio's regardless of the
aggregation size limit.
Analysis sponsored by Intel Corp.
OpenZFS-issue: https://www.illumos.org/issues/8005
OpenZFS-commit: https://github.com/openzfs/openzfs/pull/321Closes#5931
Enable picky cstyle checks and resolve the new warnings. The vast
majority of the changes needed were to handle minor issues with
whitespace formatting. This patch contains no functional changes.
Non-whitespace changes are as follows:
* 8 times ; to { } in for/while loop
* fix missing ; in cmd/zed/agents/zfs_diagnosis.c
* comment (confim -> confirm)
* change endline , to ; in cmd/zpool/zpool_main.c
* a number of /* BEGIN CSTYLED */ /* END CSTYLED */ blocks
* /* CSTYLED */ markers
* change == 0 to !
* ulong to unsigned long in module/zfs/dsl_scan.c
* rearrangement of module_param lines in module/zfs/metaslab.c
* add { } block around statement after for_each_online_node
Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov>
Reviewed-by: Håkan Johansson <f96hajo@chalmers.se>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#5465
OpenZFS 7090 - zfs should throttle allocations
Authored by: George Wilson <george.wilson@delphix.com>
Reviewed by: Alex Reece <alex@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: Dan Kimmel <dan.kimmel@delphix.com>
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: Paul Dagnelie <paul.dagnelie@delphix.com>
Reviewed by: Prakash Surya <prakash.surya@delphix.com>
Reviewed by: Sebastien Roy <sebastien.roy@delphix.com>
Approved by: Matthew Ahrens <mahrens@delphix.com>
Ported-by: Don Brady <don.brady@intel.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
When write I/Os are issued, they are issued in block order but the ZIO
pipeline will drive them asynchronously through the allocation stage
which can result in blocks being allocated out-of-order. It would be
nice to preserve as much of the logical order as possible.
In addition, the allocations are equally scattered across all top-level
VDEVs but not all top-level VDEVs are created equally. The pipeline
should be able to detect devices that are more capable of handling
allocations and should allocate more blocks to those devices. This
allows for dynamic allocation distribution when devices are imbalanced
as fuller devices will tend to be slower than empty devices.
The change includes a new pool-wide allocation queue which would
throttle and order allocations in the ZIO pipeline. The queue would be
ordered by issued time and offset and would provide an initial amount of
allocation of work to each top-level vdev. The allocation logic utilizes
a reservation system to reserve allocations that will be performed by
the allocator. Once an allocation is successfully completed it's
scheduled on a given top-level vdev. Each top-level vdev maintains a
maximum number of allocations that it can handle (mg_alloc_queue_depth).
The pool-wide reserved allocations (top-levels * mg_alloc_queue_depth)
are distributed across the top-level vdevs metaslab groups and round
robin across all eligible metaslab groups to distribute the work. As
top-levels complete their work, they receive additional work from the
pool-wide allocation queue until the allocation queue is emptied.
OpenZFS-issue: https://www.illumos.org/issues/7090
OpenZFS-commit: https://github.com/openzfs/openzfs/commit/4756c3d7Closes#5258
Porting Notes:
- Maintained minimal stack in zio_done
- Preserve linux-specific io sizes in zio_write_compress
- Added module params and documentation
- Updated to use optimize AVL cmp macros
perf: 2.75x faster ddt_entry_compare()
First 256bits of ddt_key_t is a block checksum, which are expected
to be close to random data. Hence, on average, comparison only needs to
look at first few bytes of the keys. To reduce number of conditional
jump instructions, the result is computed as: sign(memcmp(k1, k2)).
Sign of an integer 'a' can be obtained as: `(0 < a) - (a < 0)` := {-1, 0, 1} ,
which is computed efficiently. Synthetic performance evaluation of
original and new algorithm over 1G random keys on 2.6GHz Intel(R) Xeon(R)
CPU E5-2660 v3:
old 6.85789 s
new 2.49089 s
perf: 2.8x faster vdev_queue_offset_compare() and vdev_queue_timestamp_compare()
Compute the result directly instead of using conditionals
perf: zfs_range_compare()
Speedup between 1.1x - 2.5x, depending on compiler version and
optimization level.
perf: spa_error_entry_compare()
`bcmp()` is not suitable for comparator use. Use `memcmp()` instead.
perf: 2.8x faster metaslab_compare() and metaslab_rangesize_compare()
perf: 2.8x faster zil_bp_compare()
perf: 2.8x faster mze_compare()
perf: faster dbuf_compare()
perf: faster compares in spa_misc
perf: 2.8x faster layout_hash_compare()
perf: 2.8x faster space_reftree_compare()
perf: libzfs: faster avl tree comparators
perf: guid_compare()
perf: dsl_deadlist_compare()
perf: perm_set_compare()
perf: 2x faster range_tree_seg_compare()
perf: faster unique_compare()
perf: faster vdev_cache _compare()
perf: faster vdev_uberblock_compare()
perf: faster fuid _compare()
perf: faster zfs_znode_hold_compare()
Signed-off-by: Gvozden Neskovic <neskovic@gmail.com>
Signed-off-by: Richard Elling <richard.elling@gmail.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#5033
Async writes triggered by a self-healing IO may be issued before the
pool finishes the process of initialization. This results in a NULL
dereference of `spa->spa_dsl_pool` in vdev_queue_max_async_writes().
George Wilson recommended addressing this issue by initializing the
passed `dsl_pool_t **` prior to dmu_objset_open_impl(). Since the
caller is passing the `spa->spa_dsl_pool` this has the effect of
ensuring it's initialized.
However, since this depends on the caller knowing they must pass
the `spa->spa_dsl_pool` an additional NULL check was added to
vdev_queue_max_async_writes(). This guards against any future
restructuring of the code which might result in dsl_pool_init()
being called differently.
Signed-off-by: GeLiXin <47034221@qq.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#4652
Reviewed by: Paul Dagnelie <pcd@delphix.com>
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Approved by: Dan McDonald <danmcd@omniti.com>
Ported by: Tony Hutter <hutter2@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
OpenZFS-issue: https://www.illumos.org/issues/6531
OpenZFS-commit: https://github.com/openzfs/openzfs/commit/97e8130
Porting notes:
- Added new IO delay tracepoints, and moved common ZIO tracepoint macros
to a new trace_common.h file.
- Used zio_delay_taskq() in place of OpenZFS's timeout_generic() function.
- Updated zinject man page
- Updated zpool_scrub test files
The existing algorithm selects a preferred leaf vdev based on offset of the zio
request modulo the number of members in the mirror. It assumes the devices are
of equal performance and that spreading the requests randomly over both drives
will be sufficient to saturate them. In practice this results in the leaf vdevs
being under utilized.
The new algorithm takes into the following additional factors:
* Load of the vdevs (number outstanding I/O requests)
* The locality of last queued I/O vs the new I/O request.
Within the locality calculation additional knowledge about the underlying vdev
is considered such as; is the device backing the vdev a rotating media device.
This results in performance increases across the board as well as significant
increases for predominantly streaming loads and for configurations which don't
have evenly performing devices.
The following are results from a setup with 3 Way Mirror with 2 x HD's and
1 x SSD from a basic test running multiple parrallel dd's.
With pre-fetch disabled (vfs.zfs.prefetch_disable=1):
== Stripe Balanced (default) ==
Read 15360MB using bs: 1048576, readers: 3, took 161 seconds @ 95 MB/s
== Load Balanced (zfslinux) ==
Read 15360MB using bs: 1048576, readers: 3, took 297 seconds @ 51 MB/s
== Load Balanced (locality freebsd) ==
Read 15360MB using bs: 1048576, readers: 3, took 54 seconds @ 284 MB/s
With pre-fetch enabled (vfs.zfs.prefetch_disable=0):
== Stripe Balanced (default) ==
Read 15360MB using bs: 1048576, readers: 3, took 91 seconds @ 168 MB/s
== Load Balanced (zfslinux) ==
Read 15360MB using bs: 1048576, readers: 3, took 108 seconds @ 142 MB/s
== Load Balanced (locality freebsd) ==
Read 15360MB using bs: 1048576, readers: 3, took 48 seconds @ 320 MB/s
In addition to the performance changes the code was also restructured, with
the help of Justin Gibbs, to provide a more logical flow which also ensures
vdevs loads are only calculated from the set of valid candidates.
The following additional sysctls where added to allow the administrator
to tune the behaviour of the load algorithm:
* vfs.zfs.vdev.mirror.rotating_inc
* vfs.zfs.vdev.mirror.rotating_seek_inc
* vfs.zfs.vdev.mirror.rotating_seek_offset
* vfs.zfs.vdev.mirror.non_rotating_inc
* vfs.zfs.vdev.mirror.non_rotating_seek_inc
These changes where based on work started by the zfsonlinux developers:
https://github.com/zfsonlinux/zfs/pull/1487
Reviewed by: gibbs, mav, will
MFC after: 2 weeks
Sponsored by: Multiplay
References:
https://github.com/freebsd/freebsd@5c7a6f5dhttps://github.com/freebsd/freebsd@31b7f68dhttps://github.com/freebsd/freebsd@e186f564
Performance Testing:
https://github.com/zfsonlinux/zfs/pull/4334#issuecomment-189057141
Porting notes:
- The tunables were adjusted to have ZoL-style names.
- The code was modified to use ZoL's vd_nonrot.
- Fixes were done to make cstyle.pl happy
- Merge conflicts were handled manually
- freebsd/freebsd@e186f564bc by my
collegue Andriy Gapon has been included. It applied perfectly, but
added a cstyle regression.
- This replaces 556011dbec entirely.
- A typo "IO'a" has been corrected to say "IO's"
- Descriptions of new tunables were added to man/man5/zfs-module-parameters.5.
Ported-by: Richard Yao <ryao@gentoo.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#4334
Update the bounds checking for zfs_vdev_aggregation_limit so that
it has a floor of zero and a maximum value of the supported block
size for the pool.
Additionally add an early return when zfs_vdev_aggregation_limit
equals zero to disable aggregation. For very fast solid state or
memory devices it may be more expensive to perform the aggregation
than to issue the IO immediately.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
This deadlock may manifest itself in slightly different ways but
at the core it is caused by a memory allocation blocking on file-
system reclaim in the zio pipeline. This is normally impossible
because zio_execute() disables filesystem reclaim by setting
PF_FSTRANS on the thread. However, kmem cache allocations may
still indirectly block on file system reclaim while holding the
critical vq->vq_lock as shown below.
To resolve this issue zio_buf_alloc_flags() is introduced which
allocation flags to be passed. This can then be used in
vdev_queue_aggregate() with KM_NOSLEEP when allocating the
aggregate IO buffer. Since aggregating the IO is purely a
performance optimization we want this to either succeed or fail
quickly. Trying too hard to allocate this memory under the
vq->vq_lock can negatively impact performance and result in
this deadlock.
* z_wr_iss
zio_vdev_io_start
vdev_queue_io -> Takes vq->vq_lock
vdev_queue_io_to_issue
vdev_queue_aggregate
zio_buf_alloc -> Waiting on spl_kmem_cache process
* z_wr_int
zio_vdev_io_done
vdev_queue_io_done
mutex_lock -> Waiting on vq->vq_lock held by z_wr_iss
* txg_sync
spa_sync
dsl_pool_sync
zio_wait -> Waiting on zio being handled by z_wr_int
* spl_kmem_cache
spl_cache_grow_work
kv_alloc
spl_vmalloc
...
evict
zpl_evict_inode
zfs_inactive
dmu_tx_wait
txg_wait_open -> Waiting on txg_sync
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
Signed-off-by: Tim Chase <tim@chase2k.com>
Closes#3808Closes#3867
Add a missing space to the zfs_vdev_sync_write_min_active module
parameter description.
Signed-off-by: loli10K <ezomori.nozomu@gmail.com>
Signed-off-by: Richard Yao <ryao@gentoo.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#3714
As part of the stack reduction effort in
50b25b2187, a zio_t containing a taskq_ent
was added to struct vdev_queue which itself is part of struct vdev.
The taskq entry should be initialized as is currently done in zio_create()
for newly-created bare zio_t object. The rationale is the same as is
described in f467b05a26.
Signed-off-by: Tim Chase <tim@chase2k.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#3709
5027 zfs large block support
Reviewed by: Alek Pinchuk <pinchuk.alek@gmail.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Josef 'Jeff' Sipek <josef.sipek@nexenta.com>
Reviewed by: Richard Elling <richard.elling@richardelling.com>
Reviewed by: Saso Kiselkov <skiselkov.ml@gmail.com>
Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov>
Approved by: Dan McDonald <danmcd@omniti.com>
References:
https://www.illumos.org/issues/5027https://github.com/illumos/illumos-gate/commit/b515258
Porting Notes:
* Included in this patch is a tiny ISP2() cleanup in zio_init() from
Illumos 5255.
* Unlike the upstream Illumos commit this patch does not impose an
arbitrary 128K block size limit on volumes. Volumes, like filesystems,
are limited by the zfs_max_recordsize=1M module option.
* By default the maximum record size is limited to 1M by the module
option zfs_max_recordsize. This value may be safely increased up to
16M which is the largest block size supported by the on-disk format.
At the moment, 1M blocks clearly offer a significant performance
improvement but the benefits of going beyond this for the majority
of workloads are less clear.
* The illumos version of this patch increased DMU_MAX_ACCESS to 32M.
This was determined not to be large enough when using 16M blocks
because the zfs_make_xattrdir() function will fail (EFBIG) when
assigning a TX. This was immediately observed under Linux because
all newly created files must have a security xattr created and
that was failing. Therefore, we've set DMU_MAX_ACCESS to 64M.
* On 32-bit platforms a hard limit of 1M is set for blocks due
to the limited virtual address space. We should be able to relax
this one the ABD patches are merged.
Ported-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#354
Reviewed by: Andriy Gapon <avg@FreeBSD.org>
Reviewed by: Will Andrews <willa@SpectraLogic.com>
Reviewed by: Matt Ahrens <mahrens@delphix.com>
Reviewed by: George Wilson <george@delphix.com>
Approved by: Robert Mustacchi <rm@joyent.com>
References:
https://www.illumos.org/issues/5313https://github.com/illumos/illumos-gate/commit/fe319232
Ported-by: DHE <git@dehacked.net>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#3280
Commit 86dd0fd added preallocated I/O buffers. This is no longer
required after the recent kmem changes designed to make our memory
allocation interfaces behave more like those found on Illumos. A
deadlock in this situation is no longer possible.
However, these allocations still have the potential to be expensive.
So a potential future optimization might be to perform then KM_NOSLEEP
so that they either succeed of fail quicky. Either case is acceptable
here because we can safely abort the aggregation.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: Dan McDonald <danmcd@omniti.com>
Approved by: Garrett D'Amore <garrett@damore.org>
References:
https://www.illumos.org/issues/4753https://github.com/illumos/illumos-gate/commit/73527f4
Comments by Matt Ahrens from the issue tracker:
When a sync task is waiting for a txg to complete, we should hurry
it along by increasing the number of outstanding async writes
(i.e. make vdev_queue_max_async_writes() return a larger number).
Initially we might just have a tunable for "minimum async writes
while a synctask is waiting" and set it to 3.
Ported-by: Tim Chase <tim@chase2k.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#2716
As part of commit e8b96c6 the search zio used by the
vdev_queue_io_to_issue() function was moved to the heap
to minimize stack usage. Functionally this is fine, but
to maximize performance it's best to minimize the number
of dynamic allocations.
To avoid this allocation temporary space for the search
zio has been reserved in the vdev_queue structure. All
access must be serialized through the vq_lock.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ned Bass <bass6@llnl.gov>
Closes#2572
The vast majority of these changes are in Linux specific code.
They are the result of not having an automated style checker to
validate the code when it was originally written. Others were
caused when the common code was slightly adjusted for Linux.
This patch contains no functional changes. It only refreshes
the code to conform to style guide.
Everyone submitting patches for inclusion upstream should now
run 'make checkstyle' and resolve any warning prior to opening
a pull request. The automated builders have been updated to
fail a build if when 'make checkstyle' detects an issue.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#1821
4045 zfs write throttle & i/o scheduler performance work
1. The ZFS i/o scheduler (vdev_queue.c) now divides i/os into 5 classes: sync
read, sync write, async read, async write, and scrub/resilver. The scheduler
issues a number of concurrent i/os from each class to the device. Once a class
has been selected, an i/o is selected from this class using either an elevator
algorithem (async, scrub classes) or FIFO (sync classes). The number of
concurrent async write i/os is tuned dynamically based on i/o load, to achieve
good sync i/o latency when there is not a high load of writes, and good write
throughput when there is. See the block comment in vdev_queue.c (reproduced
below) for more details.
2. The write throttle (dsl_pool_tempreserve_space() and
txg_constrain_throughput()) is rewritten to produce much more consistent delays
when under constant load. The new write throttle is based on the amount of
dirty data, rather than guesses about future performance of the system. When
there is a lot of dirty data, each transaction (e.g. write() syscall) will be
delayed by the same small amount. This eliminates the "brick wall of wait"
that the old write throttle could hit, causing all transactions to wait several
seconds until the next txg opens. One of the keys to the new write throttle is
decrementing the amount of dirty data as i/o completes, rather than at the end
of spa_sync(). Note that the write throttle is only applied once the i/o
scheduler is issuing the maximum number of outstanding async writes. See the
block comments in dsl_pool.c and above dmu_tx_delay() (reproduced below) for
more details.
This diff has several other effects, including:
* the commonly-tuned global variable zfs_vdev_max_pending has been removed;
use per-class zfs_vdev_*_max_active values or zfs_vdev_max_active instead.
* the size of each txg (meaning the amount of dirty data written, and thus the
time it takes to write out) is now controlled differently. There is no longer
an explicit time goal; the primary determinant is amount of dirty data.
Systems that are under light or medium load will now often see that a txg is
always syncing, but the impact to performance (e.g. read latency) is minimal.
Tune zfs_dirty_data_max and zfs_dirty_data_sync to control this.
* zio_taskq_batch_pct = 75 -- Only use 75% of all CPUs for compression,
checksum, etc. This improves latency by not allowing these CPU-intensive tasks
to consume all CPU (on machines with at least 4 CPU's; the percentage is
rounded up).
--matt
APPENDIX: problems with the current i/o scheduler
The current ZFS i/o scheduler (vdev_queue.c) is deadline based. The problem
with this is that if there are always i/os pending, then certain classes of
i/os can see very long delays.
For example, if there are always synchronous reads outstanding, then no async
writes will be serviced until they become "past due". One symptom of this
situation is that each pass of the txg sync takes at least several seconds
(typically 3 seconds).
If many i/os become "past due" (their deadline is in the past), then we must
service all of these overdue i/os before any new i/os. This happens when we
enqueue a batch of async writes for the txg sync, with deadlines 2.5 seconds in
the future. If we can't complete all the i/os in 2.5 seconds (e.g. because
there were always reads pending), then these i/os will become past due. Now we
must service all the "async" writes (which could be hundreds of megabytes)
before we service any reads, introducing considerable latency to synchronous
i/os (reads or ZIL writes).
Notes on porting to ZFS on Linux:
- zio_t gained new members io_physdone and io_phys_children. Because
object caches in the Linux port call the constructor only once at
allocation time, objects may contain residual data when retrieved
from the cache. Therefore zio_create() was updated to zero out the two
new fields.
- vdev_mirror_pending() relied on the depth of the per-vdev pending queue
(vq->vq_pending_tree) to select the least-busy leaf vdev to read from.
This tree has been replaced by vq->vq_active_tree which is now used
for the same purpose.
- vdev_queue_init() used the value of zfs_vdev_max_pending to determine
the number of vdev I/O buffers to pre-allocate. That global no longer
exists, so we instead use the sum of the *_max_active values for each of
the five I/O classes described above.
- The Illumos implementation of dmu_tx_delay() delays a transaction by
sleeping in condition variable embedded in the thread
(curthread->t_delay_cv). We do not have an equivalent CV to use in
Linux, so this change replaced the delay logic with a wrapper called
zfs_sleep_until(). This wrapper could be adopted upstream and in other
downstream ports to abstract away operating system-specific delay logic.
- These tunables are added as module parameters, and descriptions added
to the zfs-module-parameters.5 man page.
spa_asize_inflation
zfs_deadman_synctime_ms
zfs_vdev_max_active
zfs_vdev_async_write_active_min_dirty_percent
zfs_vdev_async_write_active_max_dirty_percent
zfs_vdev_async_read_max_active
zfs_vdev_async_read_min_active
zfs_vdev_async_write_max_active
zfs_vdev_async_write_min_active
zfs_vdev_scrub_max_active
zfs_vdev_scrub_min_active
zfs_vdev_sync_read_max_active
zfs_vdev_sync_read_min_active
zfs_vdev_sync_write_max_active
zfs_vdev_sync_write_min_active
zfs_dirty_data_max_percent
zfs_delay_min_dirty_percent
zfs_dirty_data_max_max_percent
zfs_dirty_data_max
zfs_dirty_data_max_max
zfs_dirty_data_sync
zfs_delay_scale
The latter four have type unsigned long, whereas they are uint64_t in
Illumos. This accommodates Linux's module_param() supported types, but
means they may overflow on 32-bit architectures.
The values zfs_dirty_data_max and zfs_dirty_data_max_max are the most
likely to overflow on 32-bit systems, since they express physical RAM
sizes in bytes. In fact, Illumos initializes zfs_dirty_data_max_max to
2^32 which does overflow. To resolve that, this port instead initializes
it in arc_init() to 25% of physical RAM, and adds the tunable
zfs_dirty_data_max_max_percent to override that percentage. While this
solution doesn't completely avoid the overflow issue, it should be a
reasonable default for most systems, and the minority of affected
systems can work around the issue by overriding the defaults.
- Fixed reversed logic in comment above zfs_delay_scale declaration.
- Clarified comments in vdev_queue.c regarding when per-queue minimums take
effect.
- Replaced dmu_tx_write_limit in the dmu_tx kstat file
with dmu_tx_dirty_delay and dmu_tx_dirty_over_max. The first counts
how many times a transaction has been delayed because the pool dirty
data has exceeded zfs_delay_min_dirty_percent. The latter counts how
many times the pool dirty data has exceeded zfs_dirty_data_max (which
we expect to never happen).
- The original patch would have regressed the bug fixed in
zfsonlinux/zfs@c418410, which prevented users from setting the
zfs_vdev_aggregation_limit tuning larger than SPA_MAXBLOCKSIZE.
A similar fix is added to vdev_queue_aggregate().
- In vdev_queue_io_to_issue(), dynamically allocate 'zio_t search' on the
heap instead of the stack. In Linux we can't afford such large
structures on the stack.
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: Ned Bass <bass6@llnl.gov>
Reviewed by: Brendan Gregg <brendan.gregg@joyent.com>
Approved by: Robert Mustacchi <rm@joyent.com>
References:
http://www.illumos.org/issues/4045illumos/illumos-gate@69962b5647
Ported-by: Ned Bass <bass6@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#1913
3742 zfs comments need cleaner, more consistent style
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Eric Schrock <eric.schrock@delphix.com>
Approved by: Christopher Siden <christopher.siden@delphix.com>
References:
https://www.illumos.org/issues/3742illumos/illumos-gate@f717074149
Ported-by: Richard Yao <ryao@gentoo.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #1775
Porting notes:
1. The change to zfs_vfsops.c was dropped because it involves
zfs_mount_label_policy, which does not exist in the Linux port.
3537 want pool io kstats
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Eric Schrock <eric.schrock@delphix.com>
Reviewed by: Sa?o Kiselkov <skiselkov.ml@gmail.com>
Reviewed by: Garrett D'Amore <garrett@damore.org>
Reviewed by: Brendan Gregg <brendan.gregg@joyent.com>
Approved by: Gordon Ross <gwr@nexenta.com>
References:
http://www.illumos.org/issues/3537illumos/illumos-gate@c3a6601
Ported by: Cyril Plisko <cyril.plisko@mountall.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Porting Notes:
1. The patch was restructured to take advantage of the existing
spa statistics infrastructure. To accomplish this the kstat
was moved in to spa->io_stats and the init/destroy code moved
to spa_stats.c.
2. The I/O kstat was simply named <pool> which conflicted with the
pool directory we had already created. Therefore it was renamed
to <pool>/io
3. An update handler was added to allow the kstat to be zeroed.
